Tracking Control of Shape Memory Alloy Artificial Wrist Joint Using Sliding Mode Control Strategy Based on RBF Neural Network

Abstract

Shape memory alloy (SMA) is a kind of smart material that can generate deformation and recovery tension with the phase transition induced by temperature difference. SMA has been gradually used in many fields like micro robots and aerospace with its excellent flexible actuating ability. However, due to the internal material properties, there is a strong saturation hysteresis nonlinearity in SMA drivers, which leads to a severe drop in output accuracy, and even instability in some special conditions. In this paper, an artificial wrist joint was built. A sliding mode controller is proposed in this paper with the internal hysteresis estimated by the RBF networks to improve control ability. The proposed control strategy can achieve desired output performance of the artificial wrist, and the effectiveness of the proposed controller is verified with the simulation tests.